A Monitoring Services for Linux.md

Description

This combination of scripts monitors the services on an linux system, if a service is down it writes a log and sends a push notification via pushover (http://pushover.net). After all tests it flushes the mailq.

Files

monitor_services.sh - Monitors the services and writes the log entry if one is down

bo.sh - small api implementation for pushover push notification service

monitor_services.cron - setup line for crontab

Setup

1. Place both .sh files in the home directory of the root user
2. allow root to execute them
3. Also copy the content of the .cron file into your crontab

Now all is setup and the system will check every 10 minutes if the services are still up.

Notice

The log file is locate in /var/log/monitor_services.log

bo.sh

#!/bin/sh
# Bash script which implements a simple version of the pushover api to send notifications
# Credits https://github.com/2bytes/Bashover

LOGGING=true # This will print the message and response to the syslog.
URL="https://api.pushover.net/1/messages.json"

APP_KEY="aPTFZVhzgiqzm2yMd6LARZN8iAU8iz" #token
USER_KEY="ujkGtmTupCGqWgfQFZXPtKDcmxKeA3" #user

TITLE=$1
PRIORITY=1
SOUND="siren"


MESSAGE=$2


if [ ${#APP_KEY} == 0 ]
then
        echo "Error: You need to supply an app token"
        exit 1;
fi

if [ ${#USER_KEY} == 0 ]
then
        echo "Error: You need to supply a user key"
        exit 1;
fi

if [ ${#MESSAGE} == 0 ]
then
        MESSAGE=$(cat)
fi

RESPONSE=`curl -s --data token=$APP_KEY --data user=$USER_KEY --data-urlencode title="$TITLE" --data sound=$SOUND -
-data priority=$PRIORITY --data-urlencode message="$MESSAGE" $URL`


if [ $LOGGING == true ]
then
        logger "PUSH"
        logger "|"
        logger "| MESSAGE: $MESSAGE"
        logger "| RESPONSE: $RESPONSE"
        logger "|___________________________________________________________________________________________"
fi

monitor_services.cron

*/10 * * * * /root/monitor_services.sh >> /var/log/monitor_services.log 2>&1

monitor_services.sh

#!/bin/sh

SERVICES="mysql
dovecot
amavis
clamav-daemon
clamav-freshclam
postfix"

SUCCESS_MSG="Restarting the service was SUCCESSFUL!"
FAILED_MSG="Restarting the service has FAILED!"

send_alert ()
{
        if [ 0 -eq $1 ]; then
                MSG="$SUCCESS_MSG"
        else
                MSG="$FAILED_MSG"
        fi;

        DATE=`date +"%Y-%m-%d %H:%M:%S"`

        echo "[$DATE] $2 - $MSG" >> /var/log/monitor_services.log

        /root/bo.sh "$2" "$MSG" &>/dev/null
}

is_service_running ()
{
#       service $1 status | grep "active (running)" > /dev/null
        `/etc/init.d/$1 status | grep "active (running)" > /dev/null`
        return $?
}


for service in $SERVICES
do
        is_service_running $service

        if [ 0 -ne $? ]; then
                `/etc/init.d/$service restart`
                is_service_running $service
                send_alert $? "$service service down!"
        fi;
done

/usr/sbin/postfix flush